Statistics and Its Interface

Volume 2 (2009)

Number 3

Two oracle inequalities for regularized boosting classifiers

Pages: 271 – 284

DOI: https://dx.doi.org/10.4310/SII.2009.v2.n3.a2

Author

Ingo Steinwart (Information Sciences Group CCS-3, Los Alamos National Laboratory, Los Alamos, New Mexico, U.S.A.)

Abstract

We derive two oracle inequalities for regularized boosting algorithms for classification. The first oracle inequality generalizes and refines a result from Blanchard et al. (2003), while the second oracle inequality leads to faster learning rates than those of Blanchard et al. (2003) whenever the set of weak learners does not perfectly approximate the target function. The techniques leading to the second oracle inequality are based on the well-known approach of adding some artificial noise to the labeling process.

Keywords

statistical learning theory, boosting, regularization, oracle inequality, learning rates

2010 Mathematics Subject Classification

Primary 68T05. Secondary 62G20, 62H30, 68Q32, 68T10.

Published 1 January 2009